I. Field of the Invention
The present invention relates to communications. More particularly, the present invention relates to a novel and improved method and apparatus for maximizing system throughput and increasing signal diversity by dynamically multiplexing signals onto multiple carriers in a spread spectrum communication system.
II. Description of the Related Art
The present invention is concerned with transmitting data at rates which are higher than the maximum data rate of a single CDMA channel. A traditional CDMA channel (as standardized for cellular communication in the United States) is capable of carry digital data at a maximum rate of 9.6 bits per second using a 64 bit Walsh spreading function at 1.2288 MHz.
Many solutions to this problem have been proposed. One solution is to allocate multiple channels to the users and allow those users to transmit and receive data in parallel on the plurality of channels available to them. Two methods for providing multiple CDMA channels for use by a single user are described in U.S. Pat. No. 6,005,855, entitled xe2x80x9cMETHOD AND APPARATUS FOR PROVIDING VARIABLE RATE DATA IN A COMMUNICATIONS SYSTEM USING STATISTICAL MULTIPLEXINGxe2x80x9d, filed Apr. 28, 1997 and U.S. Pat. No. 5,777,990, entitled xe2x80x9cMETHOD AND APPARATUS FOR PROVIDING VARIABLE RATE DATA IN A COMMUNICATIONS SYSTEM USING NON-ORTHOGONAL OVERFLOW CHANNELSxe2x80x9d, filed Apr. 16, 1997, both of which are assigned to the assignee of the present invention and are incorporated by reference herein. In addition, frequency diversity can be obtained by transmitting data over multiple spread spectrum channels that are separated from one another in frequency. A method and apparatus for redundantly transmitting data over multiple CDMA channels is described in U.S. Pat. No. 5,166,951, entitled xe2x80x9cHIGH CAPACITY SPREAD SPECTRUM CHANNELxe2x80x9d, which is incorporated by reference herein.
The use of code division multiple access (CDMA) modulation techniques is one of several techniques for facilitating communications in which a large number of system users are present. Other multiple access communication system techniques, such as time division multiple access (TDMA), frequency division multiple access (FDMA) and AM modulation schemes such as amplitude companded single sideband (ACSSB) are known in the art. However, the spread spectrum modulation technique of CDMA has significant advantages over these other modulation techniques for multiple access communication systems.
The use of CDMA techniques in a multiple access communication system is disclosed in U.S. Pat. No. 4,901,307, entitled xe2x80x9cSPREAD SPECTRUM MULTIPLE ACCESS COMMUNICATION SYSTEM USING SATELLITE OR TERRESTRIAL REPEATERSxe2x80x9d, assigned to the assignee of the present invention and incorporated by reference herein. The use of CDMA techniques in a multiple access communication system is further disclosed in U.S. Pat. No. 5,103,459, entitled xe2x80x9cSYSTEM AND METHOD FOR GENERATING SIGNAL WAVEFORMS IN A CDMA CELLULAR TELEPHONE SYSTEMxe2x80x9d, assigned to the assignee of the present invention and incorporated by reference herein. Code division multiple access communications systems have been standardized in the United States in Telecommunications Industry Association Interim Standard IS-95, entitled xe2x80x9cMobile Station-Base Station Compatibility Standard for Dual Mode Wideband Spread Spectrum Cellular Systemxe2x80x9d, which is incorporated by reference herein.
The CDMA waveform by its inherent nature of being a wideband signal offers a form of frequency diversity by spreading the signal energy over a wide bandwidth. Therefore, frequency selective fading affects only a small part of the CDMA signal bandwidth. Space or path diversity on the forward/reverse link is obtained by providing multiple signal paths through simultaneous links to/from a mobile user through two or more antennas, cell sectors or cell-sites. Furthermore, path diversity may be obtained by exploiting the multipath environment through spread spectrum processing by allowing a signal arriving with different propagation delays to be received and processed separately. Examples of the utilization of path diversity are illustrated in copending U.S. Pat. No. 5,101,501 entitled xe2x80x9cSOFT HANDOFF IN A CDMA CELLULAR TELEPHONE SYSTEMxe2x80x9d, and U.S. Pat. No. 5,109,390 entitled xe2x80x9cDIVERSITY RECEIVER IN A CDMA CELLULAR TELEPHONE SYSTEMxe2x80x9d, both assigned to the assignee of the present invention and incorporated by reference herein.
FIG. 1 illustrates a transmission scheme for a multiple-carrier code division multiple access (CDMA) system in which each carrier carries a fixed fraction of the transmitted data. Variable rate frame of information bits are provided to encoder 2 which encodes the bits in accordance with a convolutional encoding format. The encoded symbols are provided to symbol repetition means 4. Symbol repetition means 4 repeats the encoded symbols so as to provide a fixed rate of symbols out of symbol repetition means 4, regardless of the rate of the information bits.
The repeated symbols are provided to block interleaver 6 rearranges the sequence in which the symbols are to be transmitted. The interleaving process, coupled with the forward error correction, provides time diversity which aids in the reception and error recovery of the transmitted signal in the face of burst errors. The interleaved symbols are provided to data scrambler 12. Data scrambler 12 multiplies each interleaved symbol by (+1 or xe2x88x921) according to a pseudonoise (PN) sequence. The pseudonoise sequence is provided by passing a long PN sequence generated by long code generator 8 at the chip rate through decimator 10 which selectively provides a subset of the chips of the long code sequence at the rate of the interleaved symbol stream.
The data from data scrambler 12 is provided to demultiplexer (DEMUX) 14. Demultiplexer 14 divides the data stream into three equal sub-streams. The first sub-stream is provided to transmission subsystem 15a, the second sub-stream to transmission subsystem 15b and the third sub-stream to transmission subsystem 15c. The subframes are provided to serial-to-parallel converters (BINARY TO 4 LEVEL) 16a-16c. The outputs of serial to parallel converters 16a-16c are quaternary symbols (2bits/symbol) to be transmitted in a QPSK modulation format
The signals from serial-to-parallel converters 16a-16c are provided to Walsh coders 18a-18c. In Walsh coders 18a-18c, the signals from converters 16a-16c is multiplied by a Walsh sequence consisting of xe2x96xa11 values. The Walsh coded data is provided to QPSK spreaders 20a-20c, which spread the data in accordance with two short PN sequences. The short PN sequence spread signals are provided to amplifiers 22a-22b which amplify the signals in accordance with a gain factor.
The system described above suffers from a plurality of drawbacks. First, because the data is to be provided in equal sub-streams on each of the carriers, the available numerology is limited to frames with a number of code symbols that will divide evenly by a factor of three. Table 1 below illustrates the limited number of possible rate sets which are available using the transmission system illustrated in FIG. 1.
As illustrated in Table 1, because the symbols are evenly distributed to the three carries, the total data rate is limited by the carrier with the least power available or requiring the highest SNR. That is the total data rate is equal to tree times the data rate of the xe2x80x9cworstxe2x80x9d link (here the worst means the one requiring the highest SNR or having the least power available). This reduces the system throughput, because the worst link""s rate is always chosen as the common rate for all three carriers, which results in under utilization of the channel resource on the two better links.
Second, frequency dependent fading can severely affect one of the frequencies while having a limited effect on the remaining frequencies. This implementation is inflexible and does not allow transmission of a frame to be provided in a way that reduces the effects of the poor channel. Third, because of frequency dependent fading, the fading will typically always affect the same groups of symbols of each frame. Fourth, were the implementation to be superimposed on a speech transmission system there is no good way to balance the loads carried on the different frequencies on a frame by frame basis in the face of variable speech activities in each frame. This results in loss in total system throughput. And fifth, for a system with only three frequency channels, with the implementation described, there is no method of separating the speech and data so as to provide the data on one frequency or set of frequencies and the speech on a different frequency or set of frequencies. This results in a loss of system throughput as mentioned above.
Therefore, there is a need felt for an improved multi-carrier CDMA communication system which offers greater flexibility in numerology and load balancing, better resolution in data rates supported, and which offers superior performance in the face of frequency dependent fading and uneven loading.
To better utilize the channel resource, it""s necessary to be able to transmit a different data rate on each carrier according to the channel condition and the available power on each channel. One way of doing this is by changing the ratio of the inverse-multiplexing on to each of the carriers. Instead of distributing the symbols with a ratio of 1:1:1, a more arbitrary ratio can be used together with different repetition schemes as long as the resulting symbol rate on each carrier is a factor of some Walsh function rate. Walsh function rate can be 1228800, 614400, 307200, . . . , 75 for Walsh function length from 1 to 16384.
Given the Walsh function length, if the symbol rate is lower than the Walsh function rate, symbol repetition is used to xe2x80x9cmatchxe2x80x9d the rate. The repetition factor can be any number, integer or fractional. It will be understood by one skilled in the art that when repetition is present, the total transmit power can be proportionately reduced to keep the code symbol energy constant. The Walsh function length may or may not be the same on the three carriers, depending on whether we need to save code channels. For example, if the supportable code symbol rate on the three channels are 153600 sps, 30720 sps and 102400 sps (for rate xc2xd coding, these correspond to data rates of 76.8 kbps, 15.36 kbps and 51.2 kbps, respectivelyxe2x80x94the total data rate is 143.36 kbps), then the inverse-multiplexing ratio will be 15:3:10.
If a Walsh function of length 8 is used for all three channels (assuming QPSK modulation with a QPSK symbol rate of 153.6 Ksps), then each code symbol is transmitted twice, ten times, and three times on the three channels, respectively. Additional time diversity can be obtained if the repeated symbols are further interleaved. In an alternative embodiment, different Walsh function lengths are used. For example, Walsh functions for the three channels in the example of above of length 16, 16 and 8 respectively can be used, with each code symbol transmitted once on the first channel, five times on the second, and three times on the third.
The above approach does not affect the encoder since it has to be able to handle the highest data rate anyway. All that is changed is the number of data octets at the encoder input. However, this approach does have an impact on the implementation of the interleaver because the interleaver will have many possible sizes (in terms of number of symbols) if all combinations of data rates on the three channels are allowed. One alternative to the above approach which mitigates this problem is to inverse-multiplex the code symbols out of the encoder to the three carriers directly and perform interleaving of repeated code symbols on each channel separately. This simplifies the numerology and reduces the number of possible interleaver sizes on each channel.